Efficient transfer of nucleic acid into a target cell is prerequisite for the success of almost any molecular biology application. The transfer of nucleic acid into various types of cells provides means to study gene function in living organisms, to express exogenous genes, or to regulate cell functions such as protein expression. Stably transferred inserts can also be used as primer binding sites in sequencing projects. In principle, the transfer can be classified as transient or stable. In the former case the transferred genetic material will eventually disappear from the target cells. Transient gene transfer typically utilizes plasmid constructions that do not replicate within the host cell. Because vector molecules that would replicate in mammalian cells are scarce, and in essence they are limited to those involving viral replicons (i.e. no plasmids available), the transient transfer strategy is in many cases the only straightforward gene transfer strategy for mammalian cells. For other types of cells, e.g. bacterial and lower eukaryotes such as yeast, replicating plasmids are available and therefore transient expression needs to be used only in certain specific situations in which some benefits can be envisioned (e.g. conditional expression).
In many cases stable gene transfer is the preferred option. For bacteria and lower eukaryotes plasmids that replicate within the cells are available. Accordingly, these DNA molecules can be used as gene delivery vehicles. However, the copy numbers of such plasmids typically exceeds one or two and therefore the transferred genes increase the gene dosage substantially. Typically used plasmids for bacteria and yeasts are present in tens or hundreds of copies. Increased gene dosage compared to normal situation is a potential source of artefactual or at least biased experimental results in many systems. Therefore, it would be advantageous to generate situations in which single-copy gene transfer (per haploid genome) would be possible.
In general, stable single-copy gene transfer can be achieved if transferred DNA can be inserted into the target cell's chromosomal DNA. Traditionally, this has been achieved by using different types of recombination reactions. In bacteria, homologous recombination and site-specific recombination are both widely used and in some cases yet less well characterized “illegitimate” recombination may be used. The choice of a method typically depends on whether a random or targeted mutation is required. While some of these methods are relatively trivial to use for a subset of the bacterial species, a general-purpose method would be more desirable.
Recombination reactions may also be used to stably transfer DNA into eukaryotic cell's chromosomal DNA. Homologous and site-specific recombination reactions produce targeted integrations, and “illegitimate” recombination generates non-targeted events. Utilization of transpositional recombination has been described for baker's yeast Saccharomyces cerevisiae (Ji et al 1993) and for fission yeast Schizosaccharomyces pombe (Behrens et al 2000). These strategies involve in vivo transposition in which the transposon is launched from within the cell itself. They utilize suitably modified transposons in combination with transposase proteins that are produced within a given cell. Similar systems, in which transposase proteins are produced within cells, are available also for other eukaryotic organisms; typical examples include Drosophila and Zebra fish (Rubin and Spradling 1982, Raz et al. 1997).
While transposition systems based on in vivo expression of the transposition machinery are relatively straightforward to use they are not an optimal choice for gene transfer for various reasons. For example, efficiency as well as the host-range may be limited, and target site selection may not be optimal. Viral systems, especially retroviral insertion methods, have been used to generate genomic insertions for animal cells. These strategies also have some disadvantageous properties. For example, immune response may be elicited as a response to virally-encoded proteins, and in general, constructing safe and efficient virus vectors and respective packaging cell lines for a given application is not necessarily a trivial task. Therefore, also for eukaryotic cells, a general-purpose random non-viral DNA insertion strategy would be desirable. Introduction of in vitro-assembled transposition complexes into the cells may be a choice. It is likely that utilization of in vitro-assembled DNA transposition complexes may be one of the most versatile systems for gene transfer. Recently, such a system for bacterial cells has been described and it utilizes chemical reactions based on transpositional DNA recombination (U.S. Pat. No. 6,159,736 and U.S. Pat. No. 6,294,385). Efficient systems are expected to provide a pool of mutants that can be used various ways to study many types of aspects of cellular life. These mutant pools are essential for studies involving whole genomes (i.e. functional genomics studies). However, a priori it is not possible to envision whether in vitro-assembled DNA transposition complexes would work when introduced into eukaryotic cells, especially if the components are derived from the prokaryota. The difference between prokaryotic and eukaryotic cells, especially the presence of nuclear membrane and packaging of eukaryotic genomic DNA into chromatin structure, may prevent the prokaryotic systems from functioning. In addition, in view of the stability and catalytic activity of the transposition complex, conditions within eukaryotic cells may be substantially different from prokaryotic cells. In addition, other unknown restriction system(s) may fight against incoming DNA and non-specific proteases may destroy assembled transposition complexes before they execute their function for integration. Furthermore, even if the transpositional reaction integrates the transposon into the genome, the ensuing 5-bp single-stranded regions (and in some cases 4-nt flanking DNA flaps) would need to be corrected by the host. Therefore, it is clear that the stability and efficiency of transposition complexes inside a eukaryotic cell cannot be predicted from the results with bacterial cells as disclosed in U.S. Pat. No. 6,159,736 and U.S. Pat. No. 6,294,385. Thus, to date there is no indication in the prior art that in vitro-assembled transposition complexes can generally be used for nucleic acid transfer into the cells of higher organisms (i.e. eukaryotes).
Bacteriophage Mu replicates its genome using DNA transposition machinery and is one of the best characterized mobile genetic elements (Mizuuchi 1992; Chaconas et al., 1996). We utilised for the present invention a bacteriophage Mu-derived in vitro transposition system that has been introduced recently (Haapa et al. 1999a). Mu transposition complex, the machinery within which the chemical steps of transposition take place, is initially assembled from four MuA transposase protein molecules that first bind to specific binding sites in the transposon ends. The 50 bp Mu right end DNA segment contains two of these binding sites (they are called R1 and R2 and each of them is 22 bp long, Savilahti et al. 1995). When two transposon ends meet, each bound by two MuA monomers, a transposition complex is formed through conformational changes. Then Mu transposition proceeds within the context of said transposition complex, i.e., protein-DNA complexes that are also called DNA transposition complexes or transpososomes (Mizuuchi 1992, Savilahti et al. 1995). Functional core of these complexes are assembled from a tetramer of MuA transposase protein and Mu-transposon-derived DNA-end-segments (i.e. transposon end sequences recognised by MuA) containing MuA binding sites. When the core complexes are formed they can react in divalent metal ion-dependent manner with any target DNA and insert the Mu end segments into the target (Savilahti et al 1995). A hallmark of Mu transposition is the generation of a 5-bp target site duplication (Allet, 1979; Kahmann and Kamp, 1979).
In the simplest case, the MuA transposase protein and a short 50 bp Mu right-end (R-end) fragment are the only macromolecular components required for transposition complex assembly and function (Savilahti et al. 1995, Savilahti and Mizuuchi 1996). Analogously, when two R-end sequences are located as inverted terminal repeats in a longer DNA molecule, transposition complexes form by synapsing the transposon ends. Target DNA in the Mu DNA in vitro transposition reaction can be linear, open circular, or supercoiled (Haapa et al. 1999a).
To date Mu in vitro transposition-based strategies have been utilized efficiently for a variety of molecular biology applications including DNA sequencing (Haapa et al. 1999a; Butterfield et al. 2002), generation of DNA constructions for gene targeting (Vilen et al., 2001), and functional analysis of plasmid and viral (HIV) genomic DNA regions (Haapa et al., 1999b, Laurent et al., 2000). Also, functional genomics studies on whole virus genomes of potato virus A and bacteriophage PRD1 have been conducted using the Mu in vitro transposition-based approaches (Kekarainen et al., 2002, Vilen et al., 2003). In addition, pentapeptide insertion mutagenesis method has been described (Taira et al., 1999). Recently, an insertional mutagenesis strategy for bacterial genomes has been developed in which the in vitro assembled functional transpososomes were delivered into various bacterial cells by electroporation (Lamberg et al., 2002).
E. coli is the natural host of bacteriophage Mu. It was first shown with E. coli that in vitro preassembled transposition complexes can be electroporated into the bacterial cells whereby they then integrate the transposon construct into the genome (Lamberg et al., 2002). The Mu transpososomes were also able to integrate transposons into the genomes of three other Gram negative bacteria tested, namely, Salmonella enterica (previously known as S. typhimurium), Erwinia carotovara, and Yersinia enterocolitica (Lamberg et al. 2002). In each of these four bacterial species the integrated transposons were flanked by a 5-bp target site duplication, a hallmark of Mu transposition, thus confirming that the integrations were generated by DNA transposition chemistry.